From: Felipe Balbi <balbi@ti.com> To: Rajendra Nayak <rnayak@ti.com> Cc: balbi@ti.com, Tony Lindgren <tony@atomide.com>, "Bedia, Vaibhav" <vaibhav.bedia@ti.com>, "linux-omap@vger.kernel.org" <linux-omap@vger.kernel.org>, "linux-arm-kernel@lists.infradead.org" <linux-arm-kernel@lists.infradead.org>, Mark Jackson <mpfj-list@newflow.co.uk>, Sourav Poddar <sourav.poddar@ti.com>, Paul Walmsley <paul@pwsan.com> Subject: Re: Boot hang regression 3.10.0-rc4 -> 3.10.0 Date: Tue, 9 Jul 2013 10:40:05 +0300 [thread overview] Message-ID: <20130709074005.GG5552@arwen.pp.htv.fi> (raw) In-Reply-To: <51DBB96E.90600@ti.com> [-- Attachment #1: Type: text/plain, Size: 2305 bytes --] Hi, On Tue, Jul 09, 2013 at 12:49:10PM +0530, Rajendra Nayak wrote: > >>>> Right, but calling serial_omap_restore_context() even when the context > >>>> is not lost, should not ideally cause an issue. > >>> > >>> it does in one condition. If context hasn't been saved before. And that > >>> can happen in the case of wrong pm runtime status for that device. > >>> > >>> Imagine the device is marked as suspended even though it's fully enabled > >>> (it hasn't been suspended by hwmod due to NO_IDLE flag). In that case > >>> your context structure is all zeroes (context has never been saved > >>> before) then when you call pm_runtime_get_sync() on probe() your > >>> ->runtime_resume() will get called, which will restore context, > >>> essentially undoing anything which was configured by u-boot. > >> > >> This could be a problem for drivers which do a save context in ->runtime_suspend() > >> but from what I see with omap serial, there is no save context done as part of > >> ->runtime_suspend. > > > > right, because context is "saved" in set_termios. probe() will get > > called much before set_termios() has a chance to run, right ? > > > > Same problem will trigger in that case. > > > > I still think patch below is necessary > > Right, I'll try some on those lines. Looks like a > pm_runtime_set_active() is done for the console in the non DT case in > omap_serial_init_port(). It seems to be missing in the DT case. yeah, based on conversation I had with Alan Stern a while back, pm_runtime_set_*() is best called by the underlying 'bus' and I tend to agree with him on that. Our HWMOD being our 'bus' (not really, but still), is the entity which knows if the device is powered or not and should make sure to call pm_runtime_set_*() for all devices. > Although I feel this should fix the issue we have right now, I wonder > if there could ever be a case with uart being suspended and having to > resume again before a set_termios? What I mean to ask is, if the omap > serial driver assuming a resume to happen only post a set_termios is > always valid. Meaning, probe() -> suspend() -> resume() ? I don't think that would ever happen. ->set_termios() would always get called and that's actually what triggers the resume. -- balbi [-- Attachment #2: Digital signature --] [-- Type: application/pgp-signature, Size: 836 bytes --]
WARNING: multiple messages have this Message-ID (diff)
From: balbi@ti.com (Felipe Balbi) To: linux-arm-kernel@lists.infradead.org Subject: Boot hang regression 3.10.0-rc4 -> 3.10.0 Date: Tue, 9 Jul 2013 10:40:05 +0300 [thread overview] Message-ID: <20130709074005.GG5552@arwen.pp.htv.fi> (raw) In-Reply-To: <51DBB96E.90600@ti.com> Hi, On Tue, Jul 09, 2013 at 12:49:10PM +0530, Rajendra Nayak wrote: > >>>> Right, but calling serial_omap_restore_context() even when the context > >>>> is not lost, should not ideally cause an issue. > >>> > >>> it does in one condition. If context hasn't been saved before. And that > >>> can happen in the case of wrong pm runtime status for that device. > >>> > >>> Imagine the device is marked as suspended even though it's fully enabled > >>> (it hasn't been suspended by hwmod due to NO_IDLE flag). In that case > >>> your context structure is all zeroes (context has never been saved > >>> before) then when you call pm_runtime_get_sync() on probe() your > >>> ->runtime_resume() will get called, which will restore context, > >>> essentially undoing anything which was configured by u-boot. > >> > >> This could be a problem for drivers which do a save context in ->runtime_suspend() > >> but from what I see with omap serial, there is no save context done as part of > >> ->runtime_suspend. > > > > right, because context is "saved" in set_termios. probe() will get > > called much before set_termios() has a chance to run, right ? > > > > Same problem will trigger in that case. > > > > I still think patch below is necessary > > Right, I'll try some on those lines. Looks like a > pm_runtime_set_active() is done for the console in the non DT case in > omap_serial_init_port(). It seems to be missing in the DT case. yeah, based on conversation I had with Alan Stern a while back, pm_runtime_set_*() is best called by the underlying 'bus' and I tend to agree with him on that. Our HWMOD being our 'bus' (not really, but still), is the entity which knows if the device is powered or not and should make sure to call pm_runtime_set_*() for all devices. > Although I feel this should fix the issue we have right now, I wonder > if there could ever be a case with uart being suspended and having to > resume again before a set_termios? What I mean to ask is, if the omap > serial driver assuming a resume to happen only post a set_termios is > always valid. Meaning, probe() -> suspend() -> resume() ? I don't think that would ever happen. ->set_termios() would always get called and that's actually what triggers the resume. -- balbi -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 836 bytes Desc: Digital signature URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20130709/b1c30c1f/attachment.sig>
next prev parent reply other threads:[~2013-07-09 7:40 UTC|newest] Thread overview: 96+ messages / expand[flat|nested] mbox.gz Atom feed top 2013-07-04 13:25 Boot hang regression 3.10.0-rc4 -> 3.10.0 Mark Jackson 2013-07-04 13:25 ` Mark Jackson 2013-07-04 15:14 ` Mark Jackson 2013-07-04 15:14 ` Mark Jackson 2013-07-04 16:00 ` Mark Jackson 2013-07-04 16:00 ` Mark Jackson 2013-07-05 8:11 ` Bedia, Vaibhav 2013-07-05 8:11 ` Bedia, Vaibhav 2013-07-05 11:59 ` Tony Lindgren 2013-07-05 11:59 ` Tony Lindgren 2013-07-05 13:20 ` Bedia, Vaibhav 2013-07-05 13:20 ` Bedia, Vaibhav 2013-07-05 13:31 ` Bedia, Vaibhav 2013-07-05 13:31 ` Bedia, Vaibhav 2013-07-08 11:25 ` Tony Lindgren 2013-07-08 11:25 ` Tony Lindgren 2013-07-08 12:16 ` Tony Lindgren 2013-07-08 12:16 ` Tony Lindgren 2013-07-08 12:41 ` Rajendra Nayak 2013-07-08 12:41 ` Rajendra Nayak 2013-07-08 13:10 ` Tony Lindgren 2013-07-08 13:10 ` Tony Lindgren 2013-07-08 13:20 ` Rajendra Nayak 2013-07-08 13:20 ` Rajendra Nayak 2013-07-08 13:25 ` Rajendra Nayak 2013-07-08 13:25 ` Rajendra Nayak 2013-07-08 13:35 ` Felipe Balbi 2013-07-08 13:35 ` Felipe Balbi 2013-07-09 5:33 ` Rajendra Nayak 2013-07-09 5:33 ` Rajendra Nayak 2013-07-09 6:42 ` Felipe Balbi 2013-07-09 6:42 ` Felipe Balbi 2013-07-09 7:19 ` Rajendra Nayak 2013-07-09 7:19 ` Rajendra Nayak 2013-07-09 7:40 ` Felipe Balbi [this message] 2013-07-09 7:40 ` Felipe Balbi 2013-07-09 18:59 ` Grygorii Strashko 2013-07-09 18:59 ` Grygorii Strashko 2013-07-09 19:41 ` Felipe Balbi 2013-07-09 19:41 ` Felipe Balbi 2013-07-10 12:16 ` Grygorii Strashko 2013-07-10 12:16 ` Grygorii Strashko 2013-07-10 12:25 ` Felipe Balbi 2013-07-10 12:25 ` Felipe Balbi 2013-07-10 8:22 ` Kevin Hilman 2013-07-10 8:22 ` Kevin Hilman 2013-07-10 12:10 ` Tony Lindgren 2013-07-10 12:10 ` Tony Lindgren 2013-07-10 12:27 ` Tony Lindgren 2013-07-10 12:27 ` Tony Lindgren 2013-07-10 14:26 ` Tony Lindgren 2013-07-10 14:26 ` Tony Lindgren 2013-07-10 16:07 ` Felipe Balbi 2013-07-10 16:07 ` Felipe Balbi 2013-07-10 16:11 ` Felipe Balbi 2013-07-10 16:11 ` Felipe Balbi 2013-07-11 6:32 ` Tony Lindgren 2013-07-11 6:32 ` Tony Lindgren 2013-07-11 9:59 ` Grygorii Strashko 2013-07-11 9:59 ` Grygorii Strashko 2013-07-12 0:40 ` Suman Anna 2013-07-12 0:40 ` Suman Anna 2013-07-15 6:44 ` Rajendra Nayak 2013-07-15 6:44 ` Rajendra Nayak 2013-07-15 10:01 ` Rajendra Nayak 2013-07-15 10:01 ` Rajendra Nayak 2013-07-15 19:23 ` Suman Anna 2013-07-15 19:23 ` Suman Anna 2013-07-16 6:30 ` Rajendra Nayak 2013-07-16 6:30 ` Rajendra Nayak 2013-07-11 9:17 ` Rajendra Nayak 2013-07-11 9:17 ` Rajendra Nayak 2013-07-11 9:26 ` Felipe Balbi 2013-07-11 9:26 ` Felipe Balbi 2013-07-11 10:16 ` [PATCH] arm: omap2plus: unidle devices which are about to probe Felipe Balbi 2013-07-11 10:16 ` Felipe Balbi 2013-07-12 11:58 ` Grygorii Strashko 2013-07-12 11:58 ` Grygorii Strashko 2013-07-12 12:10 ` Felipe Balbi 2013-07-12 12:10 ` Felipe Balbi 2013-07-12 12:27 ` Rajendra Nayak 2013-07-12 12:27 ` Rajendra Nayak 2013-07-13 22:21 ` Kevin Hilman 2013-07-13 22:21 ` Kevin Hilman 2013-07-11 9:59 ` Boot hang regression 3.10.0-rc4 -> 3.10.0 Grygorii Strashko 2013-07-11 9:59 ` Grygorii Strashko 2013-07-16 10:27 ` Grygorii Strashko 2013-07-16 10:27 ` Grygorii Strashko 2013-07-17 7:10 ` Rajendra Nayak 2013-07-17 7:10 ` Rajendra Nayak 2013-07-11 6:18 ` Rajendra Nayak 2013-07-11 6:18 ` Rajendra Nayak 2013-07-11 6:24 ` Tony Lindgren 2013-07-11 6:24 ` Tony Lindgren 2013-07-11 9:11 ` Rajendra Nayak 2013-07-11 9:11 ` Rajendra Nayak
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20130709074005.GG5552@arwen.pp.htv.fi \ --to=balbi@ti.com \ --cc=linux-arm-kernel@lists.infradead.org \ --cc=linux-omap@vger.kernel.org \ --cc=mpfj-list@newflow.co.uk \ --cc=paul@pwsan.com \ --cc=rnayak@ti.com \ --cc=sourav.poddar@ti.com \ --cc=tony@atomide.com \ --cc=vaibhav.bedia@ti.com \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.